A research proposal to evaluate the quality of ambulatory medical care in Metropolitan Detroit is described. The methodology for evaluation is a retrospective review of observed ambulatory medical records compared to the optimal medical care criteria developed by representatives of the medical providers studied. The data obtained will be used in a design to evaluate the most effective non-punitive, methods of improving ambulatory medical care. Sufficient time for interaction between study staff and providers with repeated (3) data collection periods over four years should enable us to document the changes and relate them to the interventions made. Additional data from provider and client questionnaires will enhance our own understanding of the dynamics of medical care in the six sites studied but also provide additional data for use in the behavioral change effort. The questionnaires will seek understanding of the attitudes of both parties as to the quality, satisfaction, accessibility, availability of the medical care delivered in each setting. A process-outcome study will be completed on a subjective basis in each of 10 diagnoses studied and on an objective basis in three of these diagnoses. The cooperation of four different groups or settings of practice increase the opportunity to relate organizational factors to the variables of provider and client characteristics.